I have a Qdrant vector database, and I have some collections created and 3 metadata in each one of the inserted Chunks, how can I add a prompt in the system, that when searching in the RAG takes into account the metadata values, it only takes into account the content of the page when searching, not the metadata. What is not clear to me is... how do I assign the metadata filter variable from the user's chat prompt? I mean, if the user in the chat asks for information about apples, I want the search in the Qdrant vector database to be performed only in the chunks that have the metadata “apples” and if the user asks for information about strawberries, I want the search to be performed in the chunks that have the metadata “strawberries”. Because what I don't want is that the vector database has all the data in the same collection without distinction and errors occur in the response. Thank you very much for your help. Thanks

Hi :) I am new here, very happy to find this group. What can I do to feed two types of searches in RAG system. 1) We all have it now, the vector search and it works fine. 2) keyword search related to the prompt, so if we use product number like WXZ2432 , it will know it need to share all the chunks it has this unique word. how can we do it?

Hoping someone can help get me on track. Built out this RAG system: https://www.youtube.com/watch?v=Kvf-Pr1LRe4&t=78s got it working. 1st hurdle behind me :) what I want to be able to do is have a user say something like .. send me a download link for 'xyz assessment form' .. and they will be given a link to download a pdf or, they are redirected to where the form can be found on the website or, if they ask for the same thing, they are asked for an email address, and the pdf is emailed to them ... I'd prefer option one Anyone able to help .. happy to jump on a zoom call etc .. if easier Cheers all I

How to perform in n8n a query in a QDRANT database to 2 or 3 different collections with n8n at the same time or same action. The Qdrant Vector Store node of n8n only offers options to pick, by list or by collection, but it only allows to choose one collection.

I wanted to ask you a question, which is the best and most recommended vector database, Qdrant, Pinecone or Supabase? I like the Qdrant option for being local, Pinecone for its simplicity and Supabase everyone tells me it is the best, but its free plan does not convince me because my project is small and every week the project is put on pause if not used... Thank you very much